DESCRIPTION: (Applicant's Abstract) The proposed research will investigate the relationships between the willingness/ability to delay gratification and substance use, abuse and dependence, as well as other risk factors associated with continuous substance use, abuse and dependence by adolescents and young adults. Through an initial cross-sectional survey of 8th-12th graders and a planned follow-up in the near future, the investigator attempts to examine factors associated with further involvement in substance use, abuse or dependence after experimental use, and to make a significant step toward a basic theory of addictive personality. The specific aims of the research include: ( 1 ) to reconcile various concepts in delay of gratification and to develop a reliable and valid instrument to measure the multi-component concept, based on a paper-pencil instrument and some behavioral observations; (2) to explore the relationships between the willingness/ability to delay gratification and the degree of substance involvement among adolescents; (3) to examine systematically the risk factors associated with further involvement in substance use, abuse and dependence, given lifetime use, particularly the roles of the willingness/ability to delay gratification; (4) to investigate the association between delay of gratification components and the known risk factors of substance use/abuse, such as stress-coping strategies, problem behavior, and harmful life-styles; and to evaluate their assumed roles in the etiology of substance use/abuse; (5) to inspect the relations between the willingness/ability to delay gratification and cessation and relapse of substance use and abuse. These aims will be achieved through a pre-test of 110 10th grader for instrument development, a school-based cross-sectional survey of 1,000 8th to 12th graders and a supplemental sample of 80-100 same-age adolescents who are seeking treatment or counseling services for substance-related problems. The correlation, ANOVA, multivariate regression and logit analyses, as well as structural equation models, will be used to examine a series of hypotheses on these relationships. Significant implications for drug prevention and treatment programs, and for the study of other health-related behaviors are anticipated.